Piperazinyl quinolines as chemosensitizers to increase fluconazole susceptibility of Candida albicans clinical isolates

Article abstract of DOI:10.1016/j.bmcl.2011.06.105

The effectiveness of the potent antifungal drug fluconazole is being compromised by the rise of drug-resistant fungal pathogens. While inhibition of Hsp90 or calcineurin can reverse drug resistance in Candida, such inhibitors also impair the homologous hu

Full text of DOI:10.1016/j.bmcl.2011.06.105

Bioorganic & Medicinal Chemistry Letters 21 (2011) 5502–5505
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Piperazinyl quinolines as chemosensitizers to increase fluconazole
susceptibility of Candida albicans clinical isolates
Willmen Youngsaye a, Benjamin Vincent b,c, Cathy L. Hartland a, Barbara J. Morgan a, , Sara J. Buhrlage a,à
,
Stephen Johnston a, Joshua A. Bittker a, Lawrence MacPherson a, Sivaraman Dandapani a, Michelle Palmer a,
Luke Whitesell b, Susan Lindquist b,d, Stuart L. Schreiber a,e, Benito Munoz a,
⇑
a Chemical Biology Platform and Probe Development Center, Broad Institute of MIT and Harvard, 7 Cambridge Center, Cambridge, MA 02142, USA
b Whitehead Institute for Biomedical Research, 9 Cambridge Center, Cambridge, MA 02142, USA
c Microbiology Graduate Program, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, MA 02139, USA
d Department of Biology and Howard Hughes Medical Institute, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, MA 02139, USA
e Howard Hughes Medical Institute, Broad Institute of Harvard and MIT, 7 Cambridge Center, Cambridge, MA 02142, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The effectiveness of the potent antifungal drug fluconazole is being compromised by the rise of drug-
resistant fungal pathogens. While inhibition of Hsp90 or calcineurin can reverse drug resistance in Can-
dida, such inhibitors also impair the homologous human host protein and fungal-selective chemosensi-
tizers remain rare. The MLPCN library was screened to identify compounds that selectively reverse
fluconazole resistance in a Candida albicans clinical isolate, while having no antifungal activity when
administered as a single agent. A piperazinyl quinoline was identified as a new small-molecule probe
(ML189) satisfying these criteria.
Received 24 May 2011
Revised 22 June 2011
Accepted 24 June 2011
Available online 30 June 2011
Keywords:
Candida albicans
Fluconazole
Ó 2011 Elsevier Ltd. All rights reserved.
Antifungal
Chemosensitizer
Molecular Libraries Probe Production
Centers Network (MLPCN)
Acquired drug resistance by medically relevant microorganisms
poses a grave threat to human health and has enormous economic
consequences.1–3 Fungal pathogens present a particular challenge
because they are eukaryotes and share many of the same mecha-
nisms that support the growth and survival of the human host cells
they infect. While contemporary antifungal medications such as
fluconazole remain effective, the usefulness of such drugs is com-
promised by either dose-limiting host toxicity or the frequent
emergence of high-grade resistance.2,3
The opportunistic fungus Candida albicans preferentially invades
immunocompromised individuals and is responsible for numerous
cutaneous, mucosal, and systemic blood-borne infections annually.4
The azole antifungal fluconazole is often prescribed to control such
infections, but fluconazole’s fungistatic nature and emerging resis-
tance are beginning to detract from its effectiveness. Typically, a dai-
ly regimen of 100 mg is sufficient to treat infections, but dosages as
high as 800 mg/day can be ineffective against fluconazole-resistant
C. albicans.5
Sensitizing C. albicans to fluconazole with small molecules is
one approach to combat the emerging resistance of this pathogen.
A limited number of small molecules have demonstrated modest
potential in this arena,6–11 and the proteins Hsp90 and calcineurin
appear integral to some of the resistance pathways utilized by Can-
dida.12 This tentative progress towards stemming the increasing
fluconazole resistance of Candida prompted us to screen the
National Institutes of Health Molecular Libraries Probe Production
Centers Network (NIH-MLPCN) compound collection with the goal
of identifying small molecules that act as fungal-selective chemo-
sensitizers and could be used to probe the various antifungal resis-
tance mechanisms of Candida.
A high throughput screen of ꢀ300,000 compounds evaluated
growth inhibition of the C. albicans clinical isolate CaCi-25 in the
presence of a sub-lethal concentration of fluconazole (Fig. 1, Pub-
Chem AID 1979).13,14 1,893 compounds exhibited >75% inhibition
when dosed at 9.5
lM, of which 622 possessed IC50 values less
⇑
Corresponding author. Tel.: +1 617 714 7225; fax: +1 617 714 8969.
E-mail address: bmunoz@broadinstitute.org (B. Munoz).
Present address: Goodwin Proctor LLP, 53 State Street, Boston, MA 02109, USA.
Present address: Cancer Biology, Dana-Farber Cancer Institute, 450 Brookline
than 1 M when tested in a dose-response assay.
l
An orthogonal screen evaluated the efficacy of these 622 hits in
combination with fluconazole against a more resistant C. albicans
clinical isolate CaCi-8,5,13 selecting for compounds that were active
à
Avenue, Boston, MA 02215, USA.
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.06.105

W. Youngsaye et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5502–5505
5503
both counter screens, and after examining these hits for chemical
tractability, the piperazinyl quinoline 1 (Fig. 1) was selected for fur-
ther investigation as a potential probe.
Primary screen
CaCi-2 growth inhibition with 8 µg/mL fluconazole
(>75% inhibition @ 9.5 µM)
A number of analogs structurally related to 1 were prepared and
evaluated for their ability to reverse fluconazole resistance in the C.
albicans test strains. Two critical intermediates 5 and 6 were pre-
pared by reacting excess piperazine with either 4-chlorobenzoic
acid or 4,7-dichloroquinoline (Scheme 1). Amide coupling of piper-
azinyl quinoline 5 with different carboxylic acids afforded analogs
7. Similarly, acylated piperazine 6 was appended to various aryl
chlorides and bromides to provide analogs 8.
1893 compounds
Repeat primary screen in dose-response format
(IC50 <1 µM)
622 compounds
Orthogonal screen
CaCi-8 growth inhibition with 8 µg/mL fluconazole
(IC50 <50 µM)
Upon their preparation, the resulting collection of analogs was
tested for their ability to increase fluconazole susceptibility in
CaCi-2 and CaCi-8. The fungi were incubated at 37 °C for 48 h with
403 compounds
the test compound and
8 lg/mL (26 lM) fluconazole before
growth inhibition was assessed by Alamar blue fluorometry. Gel-
danamycin, a non-selective Hsp90 inhibitor, was used as a control
Counter screen 1
Fibroblast toxicity
(IC50 >26 µM)
Counter screen 2
CaCi-2 growth inhibition
without fluconazole
(IC50 >26 µM)
for growth inhibition (100% inhibition at 10 l
M).12 The different
analogs were also screened against mammalian fibroblasts and
CaCi-2 in the absence of fluconazole to identify substances with
intrinsic toxicity or antifungal effects. In the absence of fluconaz-
ole, none of the analogs showed any appreciable activity against
296 compounds
CaCi-2 (IC50 = 15–26
(IC50 = 21–26 M).
lM) and were non-toxic to fibroblasts as well
l
O
The chemosensitizing properties of select 4-chlorobenzamide
analogs 7 on CaCi-2 and CaCi-8 are presented in Table 1. The initial
hit 1 proved to be an effective chemosensitizer of both fungal
Cl
N
N
Cl
strains (IC50 = 0.7 and 1.3 lM, respectively) but suffered from poor
N
solubility. Attempts to incorporate alternative para-substituents
typically lowered the effectiveness of the resulting compounds as
demonstrated with 7a–c. Similarly, relocation of the chloro substi-
tuent to the meta- or ortho-positions (e.g., 7d–e) was not tolerated,
although the 3,4-dichloro variant 7f displayed levels of activity
comparable to the original hit.
1
Figure 1. Summary of HTS campaign of the MLPCN ꢀ300,000 compound collection.
The selection criterion for each assay is given in parentheses.
with IC50’s below 50 lM. At this stage, 403 compounds were iden-
tified as chemosensitizers of both CaCi-2 and CaCi-8.
Amides derived from alkanoic acids were significantly more sol-
uble than 1 but were less effective chemosensitizers. Analogs bear-
ing linear (7h), branched (7i), and cyclic (7j–k) alkyl chains were
partial growth inhibitors of CaCi-2 at micromolar concentrations.
However, the cycloalkane amides 7j–k displayed greater effective-
ness against the more resistant CaCi-8 strain than their acyclic
counterparts (7h–i). Ureas such as 7l were prepared from reacting
To remove hits with undesirable activity profiles, two counter
screens were incorporated into the late stages of the screening
campaign. The first employed murine 3T3 fibroblasts to assay
non-selective mammalian cell toxicity, while re-evaluating
CaCi-2 in the absence of fluconazole identified inherently
fungitoxic substances. 296 of 403 candidates successfully passed
Cl
COOH
N
Cl
Cl
Cl
O
HN
3
4
NH
N
N
HN
(a)
(b)
NH
N
Cl
2
5
6
(c) or (d)
(e) or (f)
O
O
Cl
R
N
N
N
N
Cl
Ar
N
8
7
R = alkyl, aryl, -NHR
Scheme 1. Synthesis of analogs. Reagents and conditions: (a) Et3N, 130 °C; (b) EDCI, DMAP, CH2Cl2; (c) R-CO2H, EDCI, DMAP, CH2Cl2; (d) RNCO, CH2Cl2; (e) ArCl, Et3N, 130 °C;
(f) ArBr, NaOt-Bu, 15 mol % BINAP, 5 mol % Pd2(dba)3, toluene, 80 °C.

5504
W. Youngsaye et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5502–5505
Table 1
6 with different isocyanates and showed only moderate growth
inhibition against CaCi-2 when dosed with fluconazole. Sulfona-
mides did not increase fluconazole susceptibility of either CaCi-2
or CaCi-8 (data not shown).
Activity of benzamide replacements in the presence of fluconazolea
O
Cl
R
N
Several heteroaromatic replacements of the 4-chlorophenyl
ring were also prepared. As observed with the alkanamides, com-
pounds such as 7m–o were more soluble in PBS solution but pos-
sessed lower potencies relative to 1. Despite its poor aqueous
solubility, the 4-chlorophenyl amide remained the most effective
functional group among the several dozen variants evaluated,
and its efficacy in the cellular assays suggests 1 may possess a high
binding affinity for its target.
Next, the role of the 7-chloroquinoline fragment was investi-
gated, and several representative analogs that were prepared are
summarized in Table 2. Initially, alternatives to chlorine were intro-
duced into the ring system, and the specific sites of substitution
were varied as well. Analogs 8a–d illustrate some of the permuta-
tions examined, but none were able to match the effectiveness of
1. In addition, this series of compounds did little to improve solubil-
ity. Ultimately, our efforts to modify the quinoline system were
unfruitful; the 7-chloro derivative still offered the greatest cellular
potency against both CaCi-2 and CaCi-8.
N
N
Compound
R
IC50
(l
M)b
CaCi-8
CaCi-2
1
0.7 0.5
1.3 1.6
3.1 2.8
16.8 9.3
4.2 2.7
Cl
7a
7b
7c
2.3 1.0
5.3 1.1
3.6 2.3
Me
MeO
F
In order to circumvent the steep SAR associated with the quin-
oline scaffold, a series of nitrogen heterocycles was explored as
surrogates. The various replacements often resulted in large in-
creases in solubility but these gains were offset by diminished
activity. Truncating the quinoline system to the analogous pyri-
dines (8e–g) essentially negated all cellular activity. Similarly, re-
lated heterocycles such as pyridazine 8h, pyrimidine 8i, and
pyrazine 8j were marginal inhibitors of C. albicans growth. Bicyclic
systems such as 8k and 8l showed weak inhibition against both
Candida strains, and this activity trend was also observed for the
smaller, monocyclic imidazole and thiazole counterparts (8m–n).
The 4-chlorobenzene derivative 8o was one of several phenyl ana-
logs prepared; compounds of this class were ineffective growth
inhibitors.
7d
7e
7f
4.4 2.1
7.3 3.2
1.0 0.6
7.8 5.6
14.1 7.2
2.1 1.8
Cl
Cl
Cl
Cl
7g
7h
7i
4.4 3.1
6.3 4.4
7.2 3.8
6.1 4.4
15.0 11.6
>26.0
Compound 1 was determined to be the most effective chemo-
sensitizer of C. albicans CaCi-2 and CaCi-8 in cellular assays with
Me
Me
IC50 values of 0.7 and 1.3
mains a liability for this compound (<5
l
M, respectively. Aqueous solubility re-
M in PBS solution), and
l
all attempts to increase solubility reduced cellular activity. 1 had
no discernible effect on CaCi-2 in the absence of fluconazole, and
Me
murine 3T3 fibroblasts showed no response up to 26 lM. To pro-
7j
3.6 2.4
1.7 1.1
5.3 4.1
4.0 2.7
vide insight into possible modes of action, 1 was evaluated in yeast
reporter assays for potential inhibitory activity against Hsp90-
dependent and calcineurin-dependent signaling pathways. Inhibi-
tion of these pathways has been linked to lower fluconazole resis-
7k
tance in Candida,12 but compound
mechanism (IC50 >26
1
did not affect either
Cl
l
M), suggesting that it may exploit an alter-
7l
4.2 2.2
2.3 2.5
N
H
native pathway to reverse fluconazole resistance in C. albicans. Tar-
get identification via resistance studies are ongoing and will
hopefully uncover novel fungal stress response pathways that
could further yield additional therapeutic targets for future gener-
ations of antifungals.
The current work evaluated over 300,000 compounds of the NIH-
MLPCN collection to identify agents capable of reversing fluconazole
resistance in clinical isolates of Candida albicans. After triage through
a series of counterscreens and an orthogonal assay, 296 hits with the
ability to sensitize two Candida albicans clinical isolates towards
fluconazole treatment were identified. Piperazinyl quinoline 1 was
selected for further SAR investigation and optimization efforts. As
a result of these synthetic studies, compound 1 emerged as the most
effective chemosensitizer, able to achieve full growth inhibition of
7m
7n
6.7 3.2
4.9 2.3
10.9 9.0
4.1 3.0
Cl
Cl
N
N
7o
1.9 0.9
4.0 3.0
S
a
CaCi-2 and CaCi-8 cells were incubated at 37 °C for 48 h with test compound
and 8 g/mL (26 M) fluconazole.
Average of at least three runs.
l
l
b

W. Youngsaye et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5502–5505
5505
Table 2
both CaCi-2 (97–100%inhibition)and the more resistant CaCi-8 (97–
100% inhibition) when dosed at 1 M in conjunction with 26
fluconazole (8 g/mL). In addition, it was determined that 1 was
Activity of select quinoline analogs in the presence of fluconazolea
l
lM
O
l
not inherently toxic or fungicidal and did not act via two established
resistance pathways. This compound has been registered with NIH
Molecular Libraries Program (probe ML189) and is available upon
request.15
N
N
Cl
Ar
Compound
Ar
IC50
(
l
M)b
CaCi-8
CaCi-2
Acknowledgments
The authors are grateful to Ms. Patti Aha and Ms. Lori Thomae
for their assistance in manuscript preparation. This work was
funded by the NIH-MLPCN program (1 U54 HG005032-1 awarded
to S.L.S.). BV, LW, and SL are grateful for funding from the NIH (1
R03 MH086456-01).
1
0.7 0.5
2.4 1.9
4.8 2.8
7.3 4.9
5.7 4.2
1.3 1.6
6.3 6.1
13.7 4.7
>26.0
N
N
N
Cl
Cl
8a
8b
8c
8d
Supplementary data
Supplementary data (experimental protocols for cellular assays
and for the preparation of 1) associated with this article can be
found, in the online version, at doi:10.1016/j.bmcl.2011.06.105.
OMe
References and notes
Me
N
1. Levy, S. B.; Marshall, B. Nat. Med. 2004, 10, S122.
2. Pfaller, M. A.; Diekema, D. J. Clin. Microbiol. Rev. 2007, 20, 133.
3. Sanglard, D.; Odds, F. C. Lancet Infect. Dis. 2002, 2, 73.
4. MacCallum, D. In New Insights in Medical Mycology; Kavanagh, K., Ed.; Springer-
Netherlands: Dordrecht, 2007; pp 99–129.
5. Redding, S.; Smith, J.; Farinacci, G.; Rinaldi, M.; Fothergill, A.; Rhine-Chalberg,
J.; Pfaller, M. Clin. Infect. Dis. 1994, 18, 240.
6. DiGirolamo, J. A.; Li, X.-C.; Jacob, M. R.; Clark, A. M.; Ferreira, D. J. Nat. Prod.
2009, 72, 1524.
7. Cernicka, J.; Kozovska, Z.; Hnatova, M.; Valachovic, M.; Hapala, I.; Riedl, Z.;
Hajós, G.; Subik, J. Int. J. Antimicrob. Agents 2007, 29, 170.
8. Gamarra, S.; Rocha, E. M.; Zhang, Y.-Q.; Park, S.; Rao, R.; Perlin, D. S. Antimicrob.
Agents Chemother. 2010, 54, 1753.
9. Guo, X.-L.; Leng, P.; Yang, Y.; Yu, L.-G.; Lou, H.-X. J. Appl. Microbiol. 2008, 104,
831.
10. Courchesne, W. E. J. Pharmacol. Exp. Ther. 2002, 300, 195.
11. Mai, A.; Rotili, D.; Massa, S.; Brosch, G.; Simonetti, G.; Passariello, C.; Palamara,
A. T. Bioorg. Med. Chem. Lett. 2007, 17, 1221.
14.7 8.2
N
N
8e
8f
>10.0
24.0 3.5
>26.0
N
7.3 4.9
7.3 4.9
8g
>26.0
N
N
N
N
8h
8i
8.5 6.0
7.1 5.8
9.0 3.5
20.6 10.8
18.1 10.9
22.2 7.7
12. Cowen, L. E.; Lindquist, S. Science 2005, 309, 2185.
Me
13. The fluconazole MIC for CaCi-2 and CaCi-8 was determined to be 2
(6.5 M) and 8 g/mL (26 M), respectively. For the CaCi-2 and CaCi-8 growth
inhibition assays, a fixed fluconazole concentration of 8 g/mL (26 M) was
lg/mL
l
l
l
l
l
N
used. In CaCi-2, this corresponded to a baseline growth inhibition of 60–80%
and was sufficient to suppress the trailing growth associated with this strain.
14. The results of the HTS campaign can be viewed free of charge at: http://
pubchem.ncbi.nlm.nih.gov/assay/assay.cgi?aid=1979.
N
N
8j
15. Compound 1 (probe ML189) is also available from commercial vendors.
N
8k
8l
4.8 2.2
5.5 4.9
11.0 9.0
N
N
10.8 13.2
S
8m
8n
6.2 5.7
6.9 5.9
>26.0
Me
N
N
N
18.9 10.5
S
8o
8.7 3.9
19.0 21.4
Cl
a
CaCi-2 and CaCi-8 cells were incubated at 37 °C for 48 h with test compound
and 8 g/mL (26 M) fluconazole.
l
l
b
Average of at least three runs.